通信何时有益？超越集体智能的谱描述
原标题：When Does Communication Help? Beyond Spectral Descriptions of Collective Intelligence
AI 摘要
该研究指出，分布式推理中通信增益的聚合描述存在两个局限：一是具有相同特征值和奇异值谱的稳定线性系统可能产生符号相反的增益，仅改变消息方向即可使准确率在65.9%到91.2%之间变化；二是社区共享偏差下，更高的平均个体准确率可能同时损害未受影响社区或降低全局投票准确率。作者提出保留任务投影的局部响应近似方法，在合成任务上以0.45个百分点的均方根误差预测多轮增益，并测试了校准约束对社区伤害的缓解效果。
正文节选
When Does Communication Help? Beyond Spectral Descriptions of Collective Intelligence Abstract Communication can bring agents into agreement while making their decisions worse. We identify two limits of aggregate descriptions of communication gain in distributed inference. First, stable linear systems with fixed evidence, network and readout can have interaction and finite-time state operators with identical eigenvalue and singular-value spectra, yet produce gains of opposite sign. Changing only